The business made no sense; the founder was the best they had met

Backing a person rather than a plan is either the oldest wisdom in early-stage investing or a way of making the evidence optional.

This is an anonymised composite, not a report about a named company or person. The situation and numbers are typical rather than reported.

Where they were

A consumer app with 180,000 registered users, about 40,000 of annual revenue and no articulated business model. The founder had built and sold a small company before, had hired three engineers away from larger employers, and answered every hard question with specific evidence rather than assertion. Asked what they had changed their mind about recently, they described a pricing experiment that had failed, what it cost, and what they concluded. The round was 1,500,000 at a 12,000,000 post-money valuation; the investor was considering 100,000 for about 0.83%.

The problem

Every business metric argued against investing. Every signal about the person argued for it. The investor's own thesis was explicitly about business model evidence, which this company did not have.

The decision

Invest on the strength of the founder despite the business failing the investor's stated criteria, decline on thesis grounds, or invest a token amount as a relationship position.

What was on the table

• Invest 100,000 on the founder
  For: At pre-product-market-fit stage the team is the only asset that persists; the plan will change regardless. This founder showed the clearest evidence of learning rate the investor had encountered that year.
  Against: It breaks the written thesis. A thesis that admits an exception for an impressive person will admit one for every impressive person, and most impressive founders still fail.
• Decline on thesis grounds
  For: The thesis exists precisely to prevent decisions made on impressions. Declining keeps the standard intact and the capital available for opportunities that fit.
  Against: Mechanical application of a thesis at the earliest stage may exclude exactly the situations where judgement about people is the only available evidence.
• Invest 25,000 as a small position
  For: Keeps a relationship and some exposure without a thesis breach large enough to matter.
  Against: A token position is a way of avoiding the decision. It neither respects the thesis nor expresses the conviction that justified considering the investment.

What they chose

The investor invested 100,000, and recorded the reasoning honestly: this was a deliberate, documented exception to their thesis, made on founder quality alone, and they expected the current business to fail. They also wrote down what would make it a mistake - if the founder did not abandon the current model within a year, the judgement about learning rate was wrong.

What happened

The founder abandoned the consumer app within eight months and rebuilt the team around an internal tool they had developed for themselves. Two years later that product had 3,000,000 of ARR and raised at a 45,000,000 valuation. The investor's stake, diluted to about 0.60%, was marked at roughly 270,000. The original business was indeed worthless; the judgement that the person would find something better was correct. The investor also noted, at review, that three similar founder-led exceptions in earlier years had produced nothing, and that one success did not validate the practice.

What transfers

• An exception to a thesis is defensible when it is deliberate, documented and rare. It is not defensible when it is discovered afterwards in a pitch you enjoyed.
• Write the falsifier even for a bet on a person. 'They will abandon this model within a year' turned a vague judgement about learning rate into something checkable.
• One success does not validate a practice. Review founder exceptions as a group, not individually, or the memorable win will quietly become the rule.
• At pre-product-market-fit, what persists is the team. What that justifies is an earlier bet, not a lower standard of evidence about the team itself.

More case studies